The Charles Stark Draper Laboratory, Inc.

Christopher Bessette

Steven Jay Spector

Michael G Moebius

Steven J Byrnes

Scott Lennox

Matthew A Sinclair

Robert D Tingley

Francis J Rogomentich

Troy B Jones

A LiDAR system includes an array of optical emitters, an objective lens optically coupling each optical emitter to a respective unique portion of a field of view, an optical switching network coupled between a laser and the array of optical emitters and a controller coupled to the optical switching network and configured to cause the optical switching network to route light from the laser to a sequence of the optical emitters according to a dynamically varying temporal pattern and to vary the temporal pattern in a way that reduces risk of eye injury from the laser light.

Beam scanning methods for improved eye safety in LiDAR systems

A LiDAR system includes an array of optical emitters, an objective lens optically coupling each optical emitter to a respective unique portion of a field of view, an optical switching network coupled between a laser and the array of optical emitters and a controller coupled to the optical switching network and configured to cause the optical switching network to route light from the laser to a sequence of the optical emitters according to a dynamically varying temporal pattern and to vary the temporal pattern based at least in part on distance to an object within the field of view. The LiDAR system scans different portions of the field of view differently, such as with different laser power levels, different revisit rates and/or different scan patterns, for example based on likelihood of detecting objects of interest in the various portions or based on likely relative importance of objects likely to be found in the various portions.

Adaptive LiDAR scanning techniques for improved frame rate and safety

In an embodiment, a method includes separating a heterogeneous radio-frequency (RF) signal, received at multiple antennas, into multiple homogenous signals. The multiple antennas have a known positional arrangement. The method further includes estimating a location relative to the known positional arrangement of the antennas of an object producing the RF signal based on phase and amplitude of each homogeneous signal as received at each of the plurality of antennas. The location can further be estimated over multiple time steps, such that velocity and acceleration can be accelerated from the estimated locations.

Autonomous collision avoidance through physical layer tracking

Growing community of inventors

Melrose, MA, United States of America

Christopher Bessette